)]}' { "log": [ { "commit": "8d7e35174d02ce76e910365acaaefc281a0b72a0", "tree": "4445375bbf8c08f8032c5a013374777931949285", "parents": [ "564601a5d12f93fdde04c6bc5b097b95e7752a46" ], "author": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed Jul 06 18:18:10 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed Jul 06 18:18:10 2005 -0700" }, "message": "[IA64] fix generic/up builds\n\nJesse Barnes provided the original version of this patch months ago, but\nother changes kept conflicting with it, so it got deferred. Greg Edwards\ndug it out of obscurity just over a week ago, and almost immediately\nanother conflicting patch appeared (Bob Picco\u0027s memory-less nodes).\n\nI\u0027ve resolved the conflicts and got it running again. CONFIG_SGI_TIOCX\nis set to \"y\" in defconfig, which causes a Tiger to not boot (oops in\ntiocx_init). But that can be resolved later ... get this in now before it\ngets stale again.\n\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "af25e94d4dcfb9608846242fabdd4e6014e5c9f0", "tree": "f311db4563cb7f101520c737965e3d2180fcda67", "parents": [ "67d340f440f389e9d56201fb7c7aaa92f262feb1" ], "author": { "name": "", "email": "", "time": "Fri Jul 01 23:27:00 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed Jul 06 15:44:55 2005 -0700" }, "message": "[IA64] Make ia64 die() preempt safe\n\nSigned-off-by: Keith Owens \u003ckaos@sgi.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "67d340f440f389e9d56201fb7c7aaa92f262feb1", "tree": "a96c26a370beb23282a561ddd936e5d0aa93adde", "parents": [ "2ba3e3e65cf182436757ba13ea8d564e2950fb56", "159f597a8bd0f1d7650d5e580c93a2666c9c26d1" ], "author": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed Jul 06 15:35:18 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed Jul 06 15:35:18 2005 -0700" }, "message": "Auto merge with /home/aegl/GIT/linus\n" }, { "commit": "2ba3e3e65cf182436757ba13ea8d564e2950fb56", "tree": "7546a5141398e8ecaa493a58a4ca8af11deaf04d", "parents": [ "7fe4c1b16854f0440939c62b8102cbf5c75e7cdc" ], "author": { "name": "Keith Owens", "email": "kaos@sgi.com", "time": "Thu Jun 30 22:53:00 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed Jul 06 15:31:15 2005 -0700" }, "message": "[IA64] restore_sigcontext is not preempt safe\n\nrestore_sigcontext calls ia64_set_local_fpu_owner() which requires that\npreempt be disabled.\n\nSigned-off-by: Keith Owens \u003ckaos@sgi.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "6772926bef3c9f0ec761b39e5702535471fff70b", "tree": "b55f1b7af51e10c54781e24b5472236323d24ee5", "parents": [ "4b1294f928d9396e45f62b1c306ac8bf9fae036b" ], "author": { "name": "Rusty Lynch", "email": "rusty.lynch@intel.com", "time": "Tue Jul 05 18:54:50 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Tue Jul 05 19:19:00 2005 -0700" }, "message": "[PATCH] kprobes: fix namespace problem and sparc64 build\n\nThe following renames arch_init, a kprobes function for performing any\narchitecture specific initialization, to arch_init_kprobes in order to\ncleanup the namespace.\n\nAlso, this patch adds arch_init_kprobes to sparc64 to fix the sparc64 kprobes\nbuild from the last return probe patch.\n\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "d18bfacff20f08aecf01bb971b110ca108eef3c7", "tree": "255f862839c593c796e609328575b611e3f56cf3", "parents": [ "a68db763af9b676590c3fe9ec3f17bf18015eb2f", "fd782a4a99d2d3e818b9465c427b10f7f027d7da" ], "author": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed Jun 29 15:21:41 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed Jun 29 15:21:41 2005 -0700" }, "message": "Auto merge with /home/aegl/GIT/linus\n" }, { "commit": "a68db763af9b676590c3fe9ec3f17bf18015eb2f", "tree": "9862af81932797f0e68f7053d253e6b6b581ea3d", "parents": [ "819c67e69c4e0062787e27dd989f5f9d00d4834e" ], "author": { "name": "Peter Chubb", "email": "peterc@gelato.unsw.edu.au", "time": "Thu Jun 23 21:14:00 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Tue Jun 28 10:01:19 2005 -0700" }, "message": "[IA64] Fix another IA64 preemption problem\n\nThere\u0027s another problem shown up by Ingo\u0027s recent patch to make\nsmp_processor_id() complain if it\u0027s called with preemption enabled.\nlocal_finish_flush_tlb_mm() calls activate_context() in a situation\nwhere it could be rescheduled to another processor. This patch\ndisables preemption around the call.\n\nSigned-off-by: Peter Chubb \u003cpeterc@gelato.unsw.edu.au\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "458f935527372499b714bf4f8e646a68bb0f52e3", "tree": "bab3265d60b3c7c3f898eeefa6c9dac264d41195", "parents": [ "e15da40176f16050c8d92b36387a49370317e276" ], "author": { "name": "David Mosberger-Tang", "email": "davidm@hpl.hp.com", "time": "Wed May 04 13:25:00 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Tue Jun 28 09:28:16 2005 -0700" }, "message": "[IA64] Speed up lfetch.fault [NULL]\n\nThis patch greatly speeds up the handling of lfetch.fault instructions\nwhich result in NaT consumption. Due to the NaT-page mapped at address\n0, this is guaranteed to happen when lfetch.fault\u0027ing a NULL pointer.\nWith this patch in place, we can even define prefetch()/prefetchw() as\nlfetch.fault without significant performance degradation. More\nimportantly, it allows compilers to be more aggressive with using\nlfetch.fault on pointers that might be NULL.\n\nSigned-off-by: David Mosberger-Tang \u003cdavidm@hpl.hp.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "66b7f8a30437b8639e798f7db8e9be1da5711efa", "tree": "288703e10bab793bc399961059ad485604539955", "parents": [ "54522b6613a03807f057fd567794a31267ef85cb" ], "author": { "name": "Mark Maule", "email": "maule@sgi.com", "time": "Mon Apr 25 13:51:00 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Tue Jun 28 09:09:06 2005 -0700" }, "message": "[IA64-SGI] pcdp: add PCDP pci interface support\n\nResend 2 with changes per Bjorn Helgaas comments. Changes from original:\n\n+ Change globals to vga_console_iobase/vga_console_membase and make them\n unconditional.\n+ Address style-related comments.\n\nPatch to extend the PCDP vga setup code to support PCI io/mem translations\nfor the legacy vga ioport and ram spaces on architectures (e.g. altix) which\nneed them.\n\nSummary of the changes:\n\ndrivers/firmware/pcdp.c\ndrivers/firmware/pcdp.h\n-----------------------\n+ add declaration for the spec-defined PCI interface struct (pcdp_if_pci)\n as well as support macros.\n\n+ extend setup_vga_console() to know about pcdp_if_pci and add a couple of\n globals to hold the io and mem translation offsets if present.\n\narch/ia64/kernel/setup.c\n------------------------\n+ tweek early_console_setup() to allow multiple early console setup routines\n to be called.\n\ninclude/asm-ia64/vga.h\n----------------------\n+ make VGA_MAP_MEM vga_console_membase aware\n\nSigned-off-by: Mark Maule \u003cmaule@sgi.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "54522b6613a03807f057fd567794a31267ef85cb", "tree": "ab0d9d47779769dc5d75810aaff21972958a681f", "parents": [ "99f95e5286df2f69edab8a04c7080d986ee4233b", "f2cbb4f01936a3e4225692e03b084b78c56d386d" ], "author": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Tue Jun 28 08:24:49 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Tue Jun 28 08:24:49 2005 -0700" }, "message": "Auto merge with /home/aegl/GIT/ia64-test\n" }, { "commit": "8644d2a42bdba2d513f71c07eaf1b6f9b718b8eb", "tree": "c43b6c2fdf1b68b66906a2de69446dcec0f9af6b", "parents": [ "1cde8a16815bd85c8137d1ea556398983c597c11", "99f95e5286df2f69edab8a04c7080d986ee4233b" ], "author": { "name": "Greg KH", "email": "greg@press.(none)", "time": "Mon Jun 27 22:07:56 2005 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Mon Jun 27 22:07:56 2005 -0700" }, "message": "Merge rsync://rsync.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6\n" }, { "commit": "0e888adc41ffc02b700ade715c182a17e766af84", "tree": "b3d745d7a292213daf107c690ea43e5589397867", "parents": [ "b1bb248a5d2230a3d8ef42199c742194a8580b15" ], "author": { "name": "Kenji Kaneshige", "email": "kaneshige.kenji@jp.fujitsu.com", "time": "Thu Apr 28 00:25:58 2005 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Mon Jun 27 21:52:44 2005 -0700" }, "message": "[PATCH] ACPI based I/O APIC hot-plug: ia64 support\n\nThis is an ia64 implementation of acpi_register_ioapic() and\nacpi_unregister_ioapic() interfaces.\n\nSigned-off-by: Kenji Kaneshige \u003ckaneshige.kenji@jp.fujitsu.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "b1bb248a5d2230a3d8ef42199c742194a8580b15", "tree": "5335d22256e1c6f755f7aff01432ed2d5d722c9b", "parents": [ "8d50e332c8bd4f4e8cc76e8ed7326aa6f18182aa" ], "author": { "name": "Kenji Kaneshige", "email": "kaneshige.kenji@jp.fujitsu.com", "time": "Thu Apr 28 00:25:58 2005 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Mon Jun 27 21:52:44 2005 -0700" }, "message": "[PATCH] ACPI based I/O APIC hot-plug: add interfaces\n\nThis patch adds the following new interfaces for I/O xAPIC\nhotplug. The implementation of these interfaces depends on each\narchitecture.\n\n o int acpi_register_ioapic(acpi_handle handle, u64 phys_addr,\n\t\t\t u32 gsi_base);\n\n This new interface is to add a new I/O xAPIC specified by\n phys_addr and gsi_base pair. phys_addr is the physical address\n to which the I/O xAPIC is mapped and gsi_base is global system\n interrupt base of the I/O xAPIC. acpi_register_ioapic returns\n 0 on success, or negative value on error.\n\n o int acpi_unregister_ioapic(acpi_handle handle, u32 gsi_base);\n\n This new interface is to remove a I/O xAPIC specified by\n gsi_base. acpi_unregister_ioapic returns 0 on success, or\n negative value on error.\n\nSigned-off-by: Kenji Kaneshige \u003ckaneshige.kenji@jp.fujitsu.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "c7b645f934e52a54af58142d91fb51f881f8ce26", "tree": "97ef8305edf17c07b7e0cc63453548e99e84b8fd", "parents": [ "a528e21c235862cc1ae50e7809eb9116dc40ea0c" ], "author": { "name": "Keshavamurthy Anil S", "email": "anil.s.keshavamurthy@intel.com", "time": "Mon Jun 27 15:17:16 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Mon Jun 27 15:23:54 2005 -0700" }, "message": "[PATCH] kprobes/ia64: refuse kprobe on ivt code\n\nNot safe to insert kprobes on IVT code.\n\nThis patch checks to see if the address on which Kprobes is being inserted is\nin ivt code and if it is in ivt code then refuse to register kprobe.\n\nSigned-off-by: Anil S Keshavamurthy \u003canil.s.keshavamurthy@intel.com\u003e\nAcked-by: David Mosberger \u003cdavidm@napali.hpl.hp.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "a528e21c235862cc1ae50e7809eb9116dc40ea0c", "tree": "a755ae70dfe85d3f55c5b42df8c05dfb247f13c8", "parents": [ "97f7943d70ff0e1e92ea627c44cfacfdae65dbc4" ], "author": { "name": "Rusty Lynch", "email": "rusty.lynch@intel.com", "time": "Mon Jun 27 15:17:15 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Mon Jun 27 15:23:53 2005 -0700" }, "message": "[PATCH] kprobes/ia64: refuse inserting kprobe on slot 1\n\nWithout the ability to atomically write 16 bytes, we can not update the\nmiddle slot of a bundle, slot 1, unless we stop the machine first. This\npatch will ensure the ability to robustly insert and remove a kprobe by\nrefusing to insert a kprobe on slot 1 until a mechanism is in place to\nsafely handle this case.\n\nSigned-off-by: Rusty Lynch \u003crusty.lynch@intel.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "9508dbfe39112813612085c00d55bacd398eddc6", "tree": "5f1cac16508284f54d7402f71bebf085c4bf2e7f", "parents": [ "ba8af12f432c4f00ddb0bc1068b57b20aac93ecf" ], "author": { "name": "Rusty Lynch", "email": "rusty.lynch@intel.com", "time": "Mon Jun 27 15:17:12 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Mon Jun 27 15:23:53 2005 -0700" }, "message": "[PATCH] Return probe redesign: ia64 specific implementation\n\nThe following patch implements function return probes for ia64 using\nthe revised design. With this new design we no longer need to do some\nof the odd hacks previous required on the last ia64 return probe port\nthat I sent out for comments.\n\nNote that this new implementation still does not resolve the problem noted\nby Keith Owens where backtrace data is lost after a return probe is hit.\n\nChanges include:\n * Addition of kretprobe_trampoline to act as a dummy function for instrumented\n functions to return to, and for the return probe infrastructure to place\n a kprobe on on, gaining control so that the return probe handler\n can be called, and so that the instruction pointer can be moved back\n to the original return address.\n * Addition of arch_init(), allowing a kprobe to be registered on\n kretprobe_trampoline\n * Addition of trampoline_probe_handler() which is used as the pre_handler\n for the kprobe inserted on kretprobe_implementation. This is the function\n that handles the details for calling the return probe handler function\n and returning control back at the original return address\n * Addition of arch_prepare_kretprobe() which is setup as the pre_handler\n for a kprobe registered at the beginning of the target function by\n kernel/kprobes.c so that a return probe instance can be setup when\n a caller enters the target function. (A return probe instance contains\n all the needed information for trampoline_probe_handler to do it\u0027s job.)\n * Hooks added to the exit path of a task so that we can cleanup any left-over\n return probe instances (i.e. if a task dies while inside a targeted function\n then the return probe instance was reserved at the beginning of the function\n but the function never returns so we need to mark the instance as unused.)\n\nSigned-off-by: Rusty Lynch \u003crusty.lynch@intel.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "22e2c507c301c3dbbcf91b4948b88f78842ee6c9", "tree": "9a97c91d1362e69703aa286021daffb8a5456f4c", "parents": [ "020f46a39eb7b99a575b9f4d105fce2b142acdf1" ], "author": { "name": "Jens Axboe", "email": "axboe@suse.de", "time": "Mon Jun 27 10:55:12 2005 +0200" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Mon Jun 27 14:33:29 2005 -0700" }, "message": "[PATCH] Update cfq io scheduler to time sliced design\n\nThis updates the CFQ io scheduler to the new time sliced design (cfq\nv3). It provides full process fairness, while giving excellent\naggregate system throughput even for many competing processes. It\nsupports io priorities, either inherited from the cpu nice value or set\ndirectly with the ioprio_get/set syscalls. The latter closely mimic\nset/getpriority.\n\nThis import is based on my latest from -mm.\n\nSigned-off-by: Jens Axboe \u003caxboe@suse.de\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "7f1867a5b3dc3034cbea403b229d65eed4a7f62e", "tree": "895eff07c6d4d1181ef0e4940e41aa0eb52cae35", "parents": [ "85d7b94981e2e919697bc235aad7367b33c3864b" ], "author": { "name": "Dinakar Guniguntala", "email": "dino@in.ibm.com", "time": "Sat Jun 25 14:57:36 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Sat Jun 25 16:24:45 2005 -0700" }, "message": "[PATCH] Dynamic sched domains: ia64 changes\n\nia64 changes similar to kernel/sched.c.\n\nSigned-off-by: Dinakar Guniguntala \u003cdino@in.ibm.com\u003e\nAcked-by: Paul Jackson \u003cpj@sgi.com\u003e\nAcked-by: Nick Piggin \u003cnickpiggin@yahoo.com.au\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "687f1661d302bc70ce906594a6d3f615ef075a50", "tree": "2895a027851322c2badcd40acf6b871698f2962b", "parents": [ "68767a0ae428801649d510d9a65bb71feed44dd1" ], "author": { "name": "Nick Piggin", "email": "nickpiggin@yahoo.com.au", "time": "Sat Jun 25 14:57:21 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Sat Jun 25 16:24:42 2005 -0700" }, "message": "[PATCH] sched: sched tuning\n\nDo some basic initial tuning.\n\nSigned-off-by: Nick Piggin \u003cnickpiggin@yahoo.com.au\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "a9fa06c26f7b7914c8cdf4d309b74df3151cc227", "tree": "df8c8e9d2af2f0232d40eb67a5e76d9ae80ed2d4", "parents": [ "a02c4cb67e4ccd5ce7a13c7f04c2fedb06c35431" ], "author": { "name": "Shaohua Li", "email": "shaohua.li@intel.com", "time": "Sat Jun 25 14:55:05 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Sat Jun 25 16:24:31 2005 -0700" }, "message": "[PATCH] set cpu_state for CPU hotplug (ia64)\n\nDead CPU notifies online CPU that it\u0027s dead using cpu_state variable.\nAfter switching to physical cpu hotplug, we forgot setting the variable.\nThis patch fixes it. Currently only __cpu_die uses it. We changed other\nlocations for consistency in case others use it.\n\nSigned-off-by: Shaohua Li \u003cshaohua.li@intel.com\u003e\nAcked-by: Ashok Raj \u003cashok.raj@intel.com\u003e\nCc: \"Luck, Tony\" \u003ctony.luck@intel.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "f370513640492641b4046bfd9a6e4714f6ae530d", "tree": "46da47197fcbb3614b51c5f1fac841bf26d5e572", "parents": [ "d92de65cab5980c16d4a1c326c1ef9a591892883" ], "author": { "name": "Zwane Mwaikambo", "email": "zwane@linuxpower.ca", "time": "Sat Jun 25 14:54:50 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Sat Jun 25 16:24:29 2005 -0700" }, "message": "[PATCH] i386 CPU hotplug\n\n(The i386 CPU hotplug patch provides infrastructure for some work which Pavel\nis doing as well as for ACPI S3 (suspend-to-RAM) work which Li Shaohua\n\u003cshaohua.li@intel.com\u003e is doing)\n\nThe following provides i386 architecture support for safely unregistering and\nregistering processors during runtime, updated for the current -mm tree. In\norder to avoid dumping cpu hotplug code into kernel/irq/* i dropped the\ncpu_online check in do_IRQ() by modifying fixup_irqs(). The difference being\nthat on cpu offline, fixup_irqs() is called before we clear the cpu from\ncpu_online_map and a long delay in order to ensure that we never have any\nqueued external interrupts on the APICs. There are additional changes to s390\nand ppc64 to account for this change.\n\n1) Add CONFIG_HOTPLUG_CPU\n2) disable local APIC timer on dead cpus.\n3) Disable preempt around irq balancing to prevent CPUs going down.\n4) Print irq stats for all possible cpus.\n5) Debugging check for interrupts on offline cpus.\n6) Hacky fixup_irqs() to redirect irqs when cpus go off/online.\n7) play_dead() for offline cpus to spin inside.\n8) Handle offline cpus set in flush_tlb_others().\n9) Grab lock earlier in smp_call_function() to prevent CPUs going down.\n10) Implement __cpu_disable() and __cpu_die().\n11) Enable local interrupts in cpu_enable() after fixup_irqs()\n12) Don\u0027t fiddle with NMI on dead cpu, but leave intact on other cpus.\n13) Program IRQ affinity whilst cpu is still in cpu_online_map on offline.\n\nSigned-off-by: Zwane Mwaikambo \u003czwane@linuxpower.ca\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "852caccc89d3883522e87a91bfa89fd9c9cfe15a", "tree": "dbbb98df18b04f7624f72d4a80731d7309038b79", "parents": [ "e539c2331414e73a5a1b79fb57369d79447c1cf8" ], "author": { "name": "Anil S Keshavamurthy", "email": "anil.s.keshavamurthy@intel.com", "time": "Thu Jun 23 00:09:40 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Thu Jun 23 09:45:25 2005 -0700" }, "message": "[PATCH] Kprobes/ia64: temporary disarming of reentrant probe\n\nThis patch includes IA64 architecture specific changes(ported form i386) to\nsupport temporary disarming on reentrancy of probes.\n\nIn case of reentrancy we single step without calling user handler.\n\nSigned-of-by: Anil S Keshavamurth \u003canil.s.keshavamurthy@intel.com\u003e\n\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "89cb14c0dd0e4a7d0315d19f449389c4d49237ee", "tree": "795e52fd30c5236c365be7a3eaedb316e5c8cad5", "parents": [ "708de8f11c2901cc49fd7725baf4a0fbd7264e73" ], "author": { "name": "Keshavamurthy Anil S", "email": "anil.s.keshavamurthy@intel.com", "time": "Thu Jun 23 00:09:35 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Thu Jun 23 09:45:24 2005 -0700" }, "message": "[PATCH] Kprobes/IA64: check jprobe break before handling\n\nOnce the jprobe instrumented function returns, it executes a jprobe_break\nwhich is a break instruction with __IA64_JPROBE_BREAK value. The current\npatch checks for this break value, before assuming that jprobe instrumented\nfunction just completed.\n\nThe previous code was not checking for this value and that was a bug.\n\nSigned-off-by: Anil S Keshavamurthy \u003canil.s.keshavamurthy@intel.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "708de8f11c2901cc49fd7725baf4a0fbd7264e73", "tree": "a78c45c153c6414908104547292c177bcf309908", "parents": [ "1674eafcbd3e3c68556cf19fbf4a2c30f7add729" ], "author": { "name": "Anil S Keshavamurthy", "email": "anil.s.keshavamurthy@intel.com", "time": "Thu Jun 23 00:09:34 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Thu Jun 23 09:45:24 2005 -0700" }, "message": "[PATCH] Kprobes IA64: safe register kprobe\n\nThe current kprobes does not yet handle register kprobes on some of the\nfollowing kind of instruction which needs to be emulated in a special way.\n\n1) mov r1\u003dip\n2) chk -- Speculation check instruction\n\nThis patch attempts to fail register_kprobes() when user tries to insert\nkprobes on the above kind of instruction.\n\nSigned-off-by: Anil S Keshavamurthy \u003canil.s.keshavamurthy@intel.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "1674eafcbd3e3c68556cf19fbf4a2c30f7add729", "tree": "d57a4698902a4bc2c2e6a08d24c42ec2a5c67706", "parents": [ "a5403183d84d419651672f1aee5ff2273d185efa" ], "author": { "name": "Anil S Keshavamurthy", "email": "anil.s.keshavamurthy@intel.com", "time": "Thu Jun 23 00:09:33 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Thu Jun 23 09:45:23 2005 -0700" }, "message": "[PATCH] Kprobes IA64: cmp ctype unc support\n\nThe current Kprobes when patching the original instruction with the break\ninstruction tries to retain the original qualifying predicate(qp), however\nfor cmp.crel.ctype where ctype \u003d\u003d unc, which is a special instruction\nalways needs to be executed irrespective of qp. Hence, if the instruction\nwe are patching is of this type, then we should not copy the original qp to\nthe break instruction, this is because we always want the break fault to\nhappen so that we can emulate the instruction.\n\nThis patch is based on the feedback given by David Mosberger\n\nSigned-off-by: Anil S Keshavamurthy \u003canil.s.keshavamurthy@intel.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "a5403183d84d419651672f1aee5ff2273d185efa", "tree": "c9bcdf302b6dfc2e7230e486791ec48785d9d78f", "parents": [ "13608d6433eb34840224ef632cc444f3eb59bc13" ], "author": { "name": "Anil S Keshavamurthy", "email": "anil.s.keshavamurthy@intel.com", "time": "Thu Jun 23 00:09:32 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Thu Jun 23 09:45:23 2005 -0700" }, "message": "[PATCH] Kprobes IA64: arch_prepare_kprobes() cleanup\n\narch_prepare_kprobes() was doing lots of functionality\nin just one single function. This patch\nattempts to clean up arch_prepare_kprobes() by moving\nspecific sub task to the following (new)functions\n1)valid_kprobe_addr() --\u003e\u003e validate the given kprobe address\n2)get_kprobe_inst(slot..)-\u003e\u003e Retrives the instruction for a given slot from the bundle\n3)prepare_break_inst() --\u003e\u003e Prepares break instruction within the bundle\n\t3a)update_kprobe_inst_flag()--\u003e\u003eUpdates the internal flags, required\n\t\t\tfor proper emulation of the instruction at later\n\t\t\tpoint in time.\n\nSigned-off-by: Anil S Keshavamurthy \u003canil.s.keshavamurthy@intel.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "13608d6433eb34840224ef632cc444f3eb59bc13", "tree": "249c668170713eac16abe94877ff71628e04b360", "parents": [ "8bc76772ad653bcaad1b0af72aafb6072ef0fa87" ], "author": { "name": "Rusty Lynch", "email": "rusty.lynch@intel.com", "time": "Thu Jun 23 00:09:31 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Thu Jun 23 09:45:23 2005 -0700" }, "message": "[PATCH] Kprobes ia64 qp fix\n\nFix a bug where a kprobe still fires when the instruction is predicated\noff. So given the p6\u003d0, and we have an instruction like:\n\n(p6) move loc1\u003d0\n\nwe should not be triggering the kprobe. This is handled by carrying over\nthe qp section of the original instruction into the break instruction.\n\nSigned-off-by: Anil S Keshavamurthy \u003canil.s.keshavamurthy@intel.com\u003e\nSigned-off-by: Rusty Lynch \u003cRusty.lynch@intel.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "8bc76772ad653bcaad1b0af72aafb6072ef0fa87", "tree": "d778f0cc2640e078338d84cdc67cda403cc47d7b", "parents": [ "cd2675bf65455a45b54228b7acc0c6a26a164cb6" ], "author": { "name": "Rusty Lynch", "email": "rusty.lynch@intel.com", "time": "Thu Jun 23 00:09:30 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Thu Jun 23 09:45:23 2005 -0700" }, "message": "[PATCH] Kprobes ia64 cleanup\n\nA cleanup of the ia64 kprobes implementation such that all of the bundle\nmanipulation logic is concentrated in arch_prepare_kprobe().\n\nWith the current design for kprobes, the arch specific code only has a\nchance to return failure inside the arch_prepare_kprobe() function.\n\nThis patch moves all of the work that was happening in arch_copy_kprobe()\nand most of the work that was happening in arch_arm_kprobe() into\narch_prepare_kprobe(). By doing this we can add further robustness checks\nin arch_arm_kprobe() and refuse to insert kprobes that will cause problems.\n\nSigned-off-by: Rusty Lynch \u003cRusty.lynch@intel.com\u003e\nSigned-off-by: Anil S Keshavamurthy \u003canil.s.keshavamurthy@intel.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "cd2675bf65455a45b54228b7acc0c6a26a164cb6", "tree": "102dd9bdf79ce8a5728dd098fb2b3a574c113186", "parents": [ "b2761dc262b428475890fffd979687051beb12ba" ], "author": { "name": "Anil S Keshavamurthy", "email": "anil.s.keshavamurthy@intel.com", "time": "Thu Jun 23 00:09:29 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Thu Jun 23 09:45:23 2005 -0700" }, "message": "[PATCH] Kprobes/IA64: support kprobe on branch/call instructions\n\nThis patch is required to support kprobe on branch/call instructions.\n\nSigned-off-by: Anil S Keshavamurthy \u003canil.s.keshavamurthy@intel.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "b2761dc262b428475890fffd979687051beb12ba", "tree": "088fcb3fb1a5e3fc73614dcbfcde9aa2ff7bc1ce", "parents": [ "fd7b231ff98578308d8f5fb76a25a369ce1074ae" ], "author": { "name": "Anil S Keshavamurthy", "email": "anil.s.keshavamurthy@intel.com", "time": "Thu Jun 23 00:09:28 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Thu Jun 23 09:45:22 2005 -0700" }, "message": "[PATCH] Kprobes/IA64: architecture specific JProbes support\n\nThis patch adds IA64 architecture specific JProbes support on top of Kprobes\n\nSigned-off-by: Anil S Keshavamurthy \u003canil.s.keshavamurthy@intel.com\u003e\nSigned-off-by: Rusty Lynch \u003cRusty.lynch@intel.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "fd7b231ff98578308d8f5fb76a25a369ce1074ae", "tree": "529490442b281ce08015376fa6247647ca232e69", "parents": [ "7213b2521889eb087eed8abaa48d1a692575da3e" ], "author": { "name": "Anil S Keshavamurthy", "email": "anil.s.keshavamurthy@intel.com", "time": "Thu Jun 23 00:09:28 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Thu Jun 23 09:45:22 2005 -0700" }, "message": "[PATCH] Kprobes/IA64: arch specific handling\n\nThis is an IA64 arch specific handling of Kprobes\n\nSigned-off-by: Anil S Keshavamurthy \u003canil.s.keshavamurthy@intel.com\u003e\nSigned-off-by: Rusty Lynch \u003cRusty.lynch@intel.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "7213b2521889eb087eed8abaa48d1a692575da3e", "tree": "8a9a0b5cfaa9824de97d9dae45b20d2a7309db5b", "parents": [ "0aa55e4d7db822059fe8132fe9f2b7773c48216c" ], "author": { "name": "Anil S Keshavamurthy", "email": "anil.s.keshavamurthy@intel.com", "time": "Thu Jun 23 00:09:27 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Thu Jun 23 09:45:22 2005 -0700" }, "message": "[PATCH] Kprobes/IA64: kdebug die notification mechanism\n\nAs many of you know that kprobes exist in the main line kernel for various\narchitecture including i386, x86_64, ppc64 and sparc64. Attached patches\nfollowing this mail are a port of Kprobes and Jprobes for IA64.\n\nI have tesed this patches for kprobes and Jprobes and this seems to work fine.\n I have tested this patch by inserting kprobes on various slots and various\ntemplates including various types of branch instructions.\n\nI have also tested this patch using the tool\nhttp://marc.theaimsgroup.com/?l\u003dlinux-kernel\u0026m\u003d111657358022586\u0026w\u003d2 and the\nkprobes for IA64 works great.\n\nHere is list of TODO things and pathes for the same will appear soon.\n\n1) Support kprobes on \"mov r1\u003dip\" type of instruction\n2) Support Kprobes and Jprobes to exist on the same address\n3) Support Return probes\n3) Architecture independent cleanup of kprobes\n\nThis patch adds the kdebug die notification mechanism needed by Kprobes.\n\nFor break instruction on Branch type slot, imm21 is ignored and value\nzero is placed in IIM register, hence we need to handle kprobes\nfor switch case zero.\n\nSigned-off-by: Anil S Keshavamurthy \u003canil.s.keshavamurthy@intel.com\u003e\nSigned-off-by: Rusty Lynch \u003cRusty.lynch@intel.com\u003e\n\nFrom: Rusty Lynch \u003crusty.lynch@intel.com\u003e\n\nAt the point in traps.c where we recieve a break with a zero value, we can\nnot say if the break was a result of a kprobe or some other debug facility.\n\nThis simple patch changes the informational string to a more correct \"break\n0\" value, and applies to the 2.6.12-rc2-mm2 tree with all the kprobes\npatches that were just recently included for the next mm cut.\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "fb7a0e36532bc231bea8adfb1dddc3961eb38940", "tree": "82cb74383809668d6d92a5124fc60d11b35c8d6e", "parents": [ "4e93d3e8859c834ee18dfd33051d24df8669d0c0", "29516d75a0b09e0a0328dd55c98a342515c9615a" ], "author": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Wed Jun 22 12:22:12 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Wed Jun 22 12:22:12 2005 -0700" }, "message": "Merge kernel.org/pub/scm/linux/kernel/git/aegl/linux-2.6.git\n\nDo arch/ia64/defconfig by hand.\n" }, { "commit": "f14f75b81187cdbe10cc53a521bf9fdf97b59f8c", "tree": "5c0d48c8a3338e6f1747e6cd55f699be96ffef1a", "parents": [ "2caaad41e4aa8f5dd999695b4ddeaa0e7f3912a4" ], "author": { "name": "Jes Sorensen", "email": "jes@wildopensource.com", "time": "Tue Jun 21 17:15:02 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Tue Jun 21 18:46:18 2005 -0700" }, "message": "[PATCH] ia64 uncached alloc\n\nThis patch contains the ia64 uncached page allocator and the generic\nallocator (genalloc). The uncached allocator was formerly part of the SN2\nmspec driver but there are several other users of it so it has been split\noff from the driver.\n\nThe generic allocator can be used by device driver to manage special memory\netc. The generic allocator is based on the allocator from the sym53c8xx_2\ndriver.\n\nVarious users on ia64 needs uncached memory. The SGI SN architecture requires\nit for inter-partition communication between partitions within a large NUMA\ncluster. The specific user for this is the XPC code. Another application is\nlarge MPI style applications which use it for synchronization, on SN this can\nbe done using special \u0027fetchop\u0027 operations but it also benefits non SN\nhardware which may use regular uncached memory for this purpose. Performance\nof doing this through uncached vs cached memory is pretty substantial. This\nis handled by the mspec driver which I will push out in a seperate patch.\n\nRather than creating a specific allocator for just uncached memory I came up\nwith genalloc which is a generic purpose allocator that can be used by device\ndrivers and other subsystems as they please. For instance to handle onboard\ndevice memory. It was derived from the sym53c7xx_2 driver\u0027s allocator which\nis also an example of a potential user (I am refraining from modifying sym2\nright now as it seems to have been under fairly heavy development recently).\n\nOn ia64 memory has various properties within a granule, ie. it isn\u0027t safe to\naccess memory as uncached within the same granule as currently has memory\naccessed in cached mode. The regular system therefore doesn\u0027t utilize memory\nin the lower granules which is mixed in with device PAL code etc. The\nuncached driver walks the EFI memmap and pulls out the spill uncached pages\nand sticks them into the uncached pool. Only after these chunks have been\nutilized, will it start converting regular cached memory into uncached memory.\nHence the reason for the EFI related code additions.\n\nSigned-off-by: Jes Sorensen \u003cjes@wildopensource.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "753ee728964e5afb80c17659cc6c3a6fd0a42fe0", "tree": "41c9a7700d0858c1f77c5bdaba97e5b636f69b06", "parents": [ "bfbb38fb808ac23ef44472d05d9bb36edfb49ed0" ], "author": { "name": "Martin Hicks", "email": "mort@sgi.com", "time": "Tue Jun 21 17:14:41 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Tue Jun 21 18:46:14 2005 -0700" }, "message": "[PATCH] VM: early zone reclaim\n\nThis is the core of the (much simplified) early reclaim. The goal of this\npatch is to reclaim some easily-freed pages from a zone before falling back\nonto another zone.\n\nOne of the major uses of this is NUMA machines. With the default allocator\nbehavior the allocator would look for memory in another zone, which might be\noff-node, before trying to reclaim from the current zone.\n\nThis adds a zone tuneable to enable early zone reclaim. It is selected on a\nper-zone basis and is turned on/off via syscall.\n\nAdding some extra throttling on the reclaim was also required (patch\n4/4). Without the machine would grind to a crawl when doing a \"make -j\"\nkernel build. Even with this patch the System Time is higher on\naverage, but it seems tolerable. Here are some numbers for kernbench\nruns on a 2-node, 4cpu, 8Gig RAM Altix in the \"make -j\" run:\n\n\t\t\twall user sys %cpu ctx sw. sleeps\n\t\t\t---- ---- --- ---- ------ ------\nNo patch\t\t1009 1384 847 258 298170 504402\nw/patch, no reclaim 880 1376 667 288 254064 396745\nw/patch \u0026 reclaim 1079 1385 926 252 291625 548873\n\nThese numbers are the average of 2 runs of 3 \"make -j\" runs done right\nafter system boot. Run-to-run variability for \"make -j\" is huge, so\nthese numbers aren\u0027t terribly useful except to seee that with reclaim\nthe benchmark still finishes in a reasonable amount of time.\n\nI also looked at the NUMA hit/miss stats for the \"make -j\" runs and the\nreclaim doesn\u0027t make any difference when the machine is thrashing away.\n\nDoing a \"make -j8\" on a single node that is filled with page cache pages\ntakes 700 seconds with reclaim turned on and 735 seconds without reclaim\n(due to remote memory accesses).\n\nThe simple zone_reclaim syscall program is at\nhttp://www.bork.org/~mort/sgi/zone_reclaim.c\n\nSigned-off-by: Martin Hicks \u003cmort@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "4ea78729b8dbfc400fe165a57b90a394a7275a54", "tree": "cc8d7def737ca5a4189481466d1e4164eef98313", "parents": [ "7b404b3459db1326bbdc7085edb1f58b98da5826" ], "author": { "name": "Matthew Chapman", "email": "matthewc@hp.com", "time": "Tue Jun 21 16:19:20 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Tue Jun 21 16:19:20 2005 -0700" }, "message": "[IA64] ptrace and restore_sigcontext() allow ar.rsc.pl\u003d\u003d0\n\nThis patch fixes handling of accesses to ar.rsc via ptrace \u0026 restore_sigcontext\n[With Thanks to Chris Wright for noticing the restore_sigcontext path]\n\nSigned-off-by: Matthew Chapman \u003cmatthewc@hp.com\u003e\nAcked-by: David Mosberger \u003cdavidm@hpl.hp.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "0393eed5c3220c9c3823a09a2d02329b8ff08b45", "tree": "c465ee4c3a0322a4e6533f4e1ec0968b8b254907", "parents": [ "52a0de2cd2d3da8f90d88e2eccb63d0cadd6ae26" ], "author": { "name": "Ken Chen", "email": "kenneth.w.chen@intel.com", "time": "Tue Jun 21 14:40:31 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Tue Jun 21 14:40:31 2005 -0700" }, "message": "[IA64] fix nested_dtlb_miss handler for hugetlb address\n\nThe nested_dtlb_miss handler currently does not handle fault from\nhugetlb address correctly. It walks the page table assuming PAGE_SIZE.\nThus when taking a fault triggered from hugetlb address, it would not\ncalculate the pgd/pmd/pte address correctly and thus result an incorrect\ninvocation of ia64_do_page_fault(). In there, kernel will signal SIGBUS\nand application dies (The faulting address is perfectly legal and we\nhave a valid pte for the corresponding user hugetlb address as well).\nThis patch fix the described kernel bug. Since nested_dtlb_miss is a\nrare event and a slow path anyway, I\u0027m making the change without #ifdef\nCONFIG_HUGETLB_PAGE for code readability. Tony, please apply.\n\nSigned-off-by: Ken Chen \u003ckenneth.w.chen@intel.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "52a0de2cd2d3da8f90d88e2eccb63d0cadd6ae26", "tree": "24d764db02eaff2882434987525d4c3ce1665e35", "parents": [ "a35f1e03b868a3179fa12344ec0d46a44fbf7574" ], "author": { "name": "Christophe Lucas", "email": "clucas@rotomalug.org", "time": "Mon Jun 20 14:50:00 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Tue Jun 21 14:21:17 2005 -0700" }, "message": "[IA64] printk needs KERN_INFO arch/ia64/kernel/smp.c\n\nprintk() calls should include appropriate KERN_* constant.\n\nSigned-off-by: Christophe Lucas \u003cclucas@rotomalug.org\u003e\nSigned-off-by: Domen Puncer \u003cdomen@coderock.org\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "34b727c135ff651f153be5757056d25678b6d018", "tree": "4bff383da7c655a1b87b10281d3a29f7d8b97835", "parents": [ "986a80d5c154808cc78170584670324a22fd8219" ], "author": { "name": "David Mosberger-Tang", "email": "davidm@hpl.hp.com", "time": "Thu Jun 09 22:36:00 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Mon Jun 20 09:34:02 2005 -0700" }, "message": "[IA64] Drop spurious paren in entry.h\n\nThe latest assembler catches this typo. (reported by Jim Wilson).\n\nSigned-off-by: David Mosberger-Tang \u003cdavidm@hpl.hp.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "f2cbb4f01936a3e4225692e03b084b78c56d386d", "tree": "f89f3d8baa250589a38a4dd2df56f84cddae3c76", "parents": [ "325a479c4c110db278ef3361460a48c4093252cc", "1016888fb69662936b32ab767c7419a3be9a69d3" ], "author": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed Jun 15 14:06:48 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed Jun 15 14:06:48 2005 -0700" }, "message": "Auto merge with /home/aegl/GIT/linus\n" }, { "commit": "a2a64769d0d3cc0380b4b6ecdcb781a2f790a69e", "tree": "387ee93cf8ee39ca64dd53380d1ea25a4f6134e7", "parents": [ "1834cd9fd27763f4e55184e4138a19ba36620174" ], "author": { "name": "Christoph Lameter", "email": "clameter@sgi.com", "time": "Thu Jun 09 12:29:00 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Thu Jun 09 13:04:30 2005 -0700" }, "message": "[IA64] Fix race condition in the rt_sigprocmask fastcall\n\ncurrent-\u003eblocked will be set to the value of current-\u003ethread_info-\u003eflags if the\ncmpxchg to update thread_info-\u003eflags fails. For performance reasons the store into\ncurrent-\u003eblocked was placed in the cmpxchg loop. However, the cmpxchg overwrites the\nregister holding the value to be stored. In the rare case of a retry the value of\nthread_info-\u003eflags will be written into current-\u003eblocked.\n\nThe fix is to use another register so that the register containing the current-\u003eblocked\nvalue is not overwritten.\n\nSigned-off-by: Christoph Lameter \u003cclameter@sgi.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "05062d96a23ec0959ee5ea969f40813170c73c0e", "tree": "9aa09b58ff455cc43b81cae10d3c4dac9fe9f462", "parents": [ "f829fd23c87918374bac0d90404fe12f0e788d52" ], "author": { "name": "Peter Chubb", "email": "peterc@gelato.unsw.edu.au", "time": "Wed Jun 08 15:50:20 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Wed Jun 08 16:21:14 2005 -0700" }, "message": "[PATCH] ia64: fix floating-point preemption problem\n\nThere\u0027ve been reports of problems with CONFIG_PREEMPT\u003dy and the high\nfloating point partition. This is caused by the possibility of preemption\nand rescheduling on a different processor while saving or restioirng the\nhigh partition.\n\nThe only places where the FPU state is touched are in ptrace, in\nswitch_to(), and where handling a floating-point exception. In switch_to()\npreemption is off. So it\u0027s only in trap.c and ptrace.c that we need to\nprevent preemption.\n\nHere is a patch that adds commentary to make the conditions clear, and adds\nappropriate preempt_{en,dis}able() calls to make it so. In trap.c I use\npreempt_enable_no_resched(), as we\u0027re about to return to user space where\nthe preemption flag will be checked anyway.\n\nSigned-off-by: Peter Chubb \u003cpeterc@gelato.unsw.edu.au\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "70aa488cff83c965c9e1850f48d82b000d0d6c1c", "tree": "9236fa880fe3fab2ee3cf2bcb0ec4a4e19ded3c5", "parents": [ "86ebacd360767f6a5cf9c8810977593dccf3f3da" ], "author": { "name": "Keith Owens", "email": "kaos@sgi.com", "time": "Fri May 27 23:09:00 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed Jun 08 12:25:24 2005 -0700" }, "message": "[IA64] Extract correct break number for break.b\n\nbreak.b does not store the break number in cr.iim, instead it stores 0,\nwhich makes all break.b instructions look like BUG(). Extract the\nbreak number from the instruction itself.\n\nSigned-off-by: Keith Owens \u003ckaos@sgi.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "86ebacd360767f6a5cf9c8810977593dccf3f3da", "tree": "9ee60c8036dcf3a260ce18580517c5a8dfe00b6b", "parents": [ "866ba633a81c1e179038f7527809d9513160a6f7" ], "author": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed Jun 08 12:12:48 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed Jun 08 12:12:48 2005 -0700" }, "message": "[IA64] Update comment to describe modes set in default control register.\n\nChristian Hildner pointed out that the comment did not match what the\ncode does in cpu_init() when we set up the default control register.\nPatch based on suggestions from Ken Chen.\n\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "866ba633a81c1e179038f7527809d9513160a6f7", "tree": "10c06d91204027a8688a2337aa061e51a3321c0f", "parents": [ "ad597bd518559f59ede8d01262cdf4467e13282e" ], "author": { "name": "Keith Owens", "email": "kaos@sgi.com", "time": "Mon Jun 06 02:04:00 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed Jun 08 11:41:31 2005 -0700" }, "message": "[IA64] Module gp must point to valid memory\n\nSome bits of the kernel assume that gp always points to valid memory,\nin particular PHYSICAL_MODE_ENTER() assumes that both gp and sp are\nvalid virtual addresses with associated physical pages. The IA64\nmodule loader puts gp well past the end of the module, with no physical\nbacking. Offsets on gp are still valid, but physical mode addressing\nbreaks for modules. Ensure that gp always falls within the module\nbody. Also ensure that gp is 8 byte aligned.\n\nSigned-off-by: Keith Owens \u003ckaos@sgi.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "b655913bf364603d17ad770dc4fb80e60555a255", "tree": "6cd3a91f7c8f8269be419e1a48d06508323900e8", "parents": [ "eff910a91ac04ab1d9e210d4f721484af3b39c8d" ], "author": { "name": "Peter Chubb", "email": "peterc@gelato.unsw.edu.au", "time": "Tue May 31 22:34:00 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed Jun 01 15:20:17 2005 -0700" }, "message": "[IA64] Cleanup compile warnings for ski config\n\nThe attached patch cleans up a compilation warning when ACPI\nis turned off (i.e., when compiling for the Ski simulator).\n\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "fffcc150a21853651ea890a605832c5cccbb6279", "tree": "e589de6d14b8efc5e538a4bd7c38452e08bdec17", "parents": [ "d11cf326bd5e785cc5a3f5a3d3f4e3a5522f4fb7" ], "author": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Tue May 31 10:38:32 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Tue May 31 10:38:32 2005 -0700" }, "message": "[IA64] Use \"PER_CPU\" form of EXPORT macro\n\nI was gently reminded that there are per-cpu forms of the EXPORT_SYMBOL macros.\n\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "d11cf326bd5e785cc5a3f5a3d3f4e3a5522f4fb7", "tree": "56912a4b84f6db3aa1d615a4174b2c2845fe870e", "parents": [ "fe12e25ebdd195a57d3fd655061fd2525609b16b" ], "author": { "name": "Zhang Yanmin", "email": "yanmin.zhang@intel.com", "time": "Sun May 22 17:47:00 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Thu May 26 10:19:07 2005 -0700" }, "message": "[IA64] sys_mmap doesn\u0027t follow posix.1 when parameter len\u003d0\n\nIn IA64 kernel, sys_mmap calls do_mmap2 and do_mmap2 returns addr if\nlen\u003d0, which means the mmap sys call succeeds.\n\nPosix.1 says:\nThe mmap() function shall fail if:\n[EINVAL] The value of len is zero. \n\nHere is a patch to fix it.\n\nSigned-off-by: Zhang Yanmin \u003cyanmin.zhang@intel.com\u003e\nAcked-by: David Mosberger \u003cdavidm@hpl.hp.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "fe12e25ebdd195a57d3fd655061fd2525609b16b", "tree": "40fd06986609b302ebac96f302881d173240c79c", "parents": [ "3d620cd75588a37c7f5547e26e2a28b046c280f0" ], "author": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed May 18 17:09:06 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed May 18 17:09:06 2005 -0700" }, "message": "[IA64] initialize spinlock pfm_alt_install_check\n\nI applied the penultimate version of the perfmon patch, which didn\u0027t have\nthe initialization of the new spinlock that was added.\n\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "a1ecf7f6e65637ba4470405ad39794710dbf85d4", "tree": "a086e2ba8bf56c398c6d50a47d91af6d977aaf63", "parents": [ "6872ec548970e9fb3ccd61013f84f9bb8b30fa9a" ], "author": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed May 18 16:06:00 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed May 18 16:14:30 2005 -0700" }, "message": "[IA64] alternate perfmon handler\n\nPatch from Charles Spirakis\n\nSome linux customers want to optimize their applications on the latest\nhardware but are not yet willing to upgrade to the latest kernel. This\npatch provides a way to plug in an alternate, basic, and GPL\u0027ed PMU\nsubsystem to help with their monitoring needs or for specialty work. It\ncan also be used in case of serious unexpected bugs in perfmon. Mutual\nexclusion between the two subsystems is guaranteed, hence no conflict\ncan arise from both subsystem being present.\n\nAcked-by: Stephane Eranian \u003ceranian@hpl.hp.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "325a479c4c110db278ef3361460a48c4093252cc", "tree": "bcfbf4d0647d9442045639a5c19da59d55190e81", "parents": [ "ebcc80c1b6629a445f7471cc1ddb48faf8a84e70", "7f9eaedf894dbaa08c157832e9a6c9c03ffed1ed" ], "author": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Tue May 17 15:53:14 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Tue May 17 15:53:14 2005 -0700" }, "message": "Merge with temp tree to get David\u0027s gdb inferior calls patch\n\n" }, { "commit": "7f9eaedf894dbaa08c157832e9a6c9c03ffed1ed", "tree": "12a561efa197fdbe78ba74b576c28f132e04660a", "parents": [ "8800cea62025a5209d110c5fa5990429239d6eee" ], "author": { "name": "David Mosberger-Tang", "email": "davidm@hpl.hp.com", "time": "Tue May 10 12:49:00 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Tue May 17 14:07:10 2005 -0700" }, "message": "[IA64] Fix convert_to_non_syscall() so gdb inferior calls work again\n\nFix convert_to_non_syscall() so it arranges for the kernel to be left\nvia ia64_leave_kernel() rather than ia64_leave_syscall(). The latter\nno longer tolerates being called with pSys\u003d0 and pNonSys\u003d1.\n\nSigned-off-by: David Mosberger-Tang \u003cdavidm@hpl.hp.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "bb68c12b40c3c745381d7ce61d7b2e371f157505", "tree": "5ff0f0f266601f5eebe82744dee9302e746c936f", "parents": [ "056b5033fb8d16095fb074fc910fad90186374f3" ], "author": { "name": "Russ Anderson", "email": "(rja@sgi.com)", "time": "Mon May 09 15:03:00 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Tue May 17 12:52:43 2005 -0700" }, "message": "[IA64-SGI] cpe interrupts are not being enabled.\n\nacpi_request_vector() is called in ia64_mca_init() to get the cpe_vector.\nThe problem is that acpi_request_vector() looks in platform_intr_list[] to \nget the vector, but platform_intr_list[] is not initialized with a valid\nvector until later (in sn_setup()). Without a valid vector the code\ndefaults to polling mode.\n\nThis patch moves the call to acpi_request_vector() from ia64_mca_init()\nto ia64_mca_late_init(), which is after platform_intr_list[] is initialized.\n\nSigned-off-by: Russ Anderson (rja@sgi.com)\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "02a017a9f37ae6842e065da919b5952eb847e869", "tree": "570c84abb8d8a23747a4ed51af3d9f574df0b93b", "parents": [ "d0dac8082cbc2b234917ec53eb591b1ddbec80bb" ], "author": { "name": "David Mosberger-Tang", "email": "davidm@hpl.hp.com", "time": "Tue May 10 11:35:00 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Tue May 17 12:33:15 2005 -0700" }, "message": "[IA64] Correct convert_to_non_syscall()\n\nconvert_to_non_syscall() has the same problem that unwind_to_user()\nused to have. Fix it likewise.\n\nSigned-off-by: David Mosberger-Tang \u003cdavidm@hpl.hp.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "bfd68594082d8384781c242aa72a7950b5cf51aa", "tree": "00b5d0612bb155a83403580a4a88a0600c91e232", "parents": [ "a84f5770170232a48e753e02ee5f0f17ee17b791" ], "author": { "name": "David Mosberger-Tang", "email": "davidm@hpl.hp.com", "time": "Wed May 04 06:42:00 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Tue May 10 13:52:00 2005 -0700" }, "message": "[IA64] Avoid .spillpsp directive in handcoded assembly\n\nSome time ago, GAS was fixed to bring the .spillpsp directive in line\nwith the Intel assembler manual (there was some disagreement as to\nwhether or not there is a built-in 16-byte offset). Unfortunately,\nthere are two places in the kernel where this directive is used in\nhandwritten assembly files and those of course relied on the \"buggy\"\nbehavior. As a result, when using a \"fixed\" assembler, the kernel\npicks up the UNaT bits from the wrong place (off by 16) and randomly\nsets NaT bits on the scratch registers. This can be noticed easily by\nlooking at a coredump and finding various scratch registers with\nunexpected NaT values. The patch below fixes this by using the\n.spillsp directive instead, which works correctly no matter what\nassembler is in use.\n\nSigned-off-by: David Mosberger-Tang \u003cdavidm@hpl.hp.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "66302f211a21bb9439a2ae7f7b6a4c386bb10ecd", "tree": "d97390da2698a9ac94e2f439c88127380bc1984b", "parents": [ "966dc11fcc48866a24d9b3fd16eeebb3f5e05931" ], "author": { "name": "David Mosberger-Tang", "email": "davidm@hpl.hp.com", "time": "Tue Apr 12 11:04:00 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Mon May 09 10:16:17 2005 -0700" }, "message": "[IA64] fix \"section mismatch\" compile-time-error\n\nI noticed this typo when trying to compile a kernel which had\nCONFIG_HOTPLUG turned off. In that case, __devinit is no longer a\nno-op and the compiler then detects a section-conflict. Fix by using\n__devinitdata instead of __devinit.\n\nSame patch also submitted by Darren Williams to fix compilation error\nusing sim_defconfig (which has CONFIG_HOTPLUG\u003dn).\n\nSigned-off-by: David Mosberger-Tang \u003cdavidm@hpl.hp.com\u003e\nSigned-off-by: Darren Williams \u003cdsw@gelato.unsw.edu.au\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "966dc11fcc48866a24d9b3fd16eeebb3f5e05931", "tree": "4c374071e47998629495f2934e80d083719090f6", "parents": [ "b3214970abbe983cd89842ae24ea00e21bba79f6" ], "author": { "name": "David Mosberger-Tang", "email": "davidm@hpl.hp.com", "time": "Fri May 06 05:25:00 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Fri May 06 10:16:07 2005 -0700" }, "message": "[IA64] Fix stack placement when INIT hits in kernel mode.\n\nWithout this patch, the stack is placed _below_ the current task\nstructure, which is risky at best.\n\nTony, I think this patch needs to go into 2.6.12, since it fixes a\nreal bug. Without it, INIT may case secondary errors, which would be\nmost unpleasant.\n\nSigned-off-by: David Mosberger-Tang \u003cdavidm@hpl.hp.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "ebcc80c1b6629a445f7471cc1ddb48faf8a84e70", "tree": "74ceae4d7d8898050ab1d8b78f70fdd05906c89d", "parents": [ "8e3e50168c8537807b7a6f78588cd72e21363262" ], "author": { "name": "David Mosberger-Tang", "email": "davidm@hpl.hp.com", "time": "Thu May 05 06:40:00 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Thu May 05 11:30:48 2005 -0700" }, "message": "[IA64] Merge audit fix for fsyscalls with syscall-optimizations\n\nSigned-off-by: David Mosberger-Tang \u003cdavidm@hpl.hp.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "bfd4bda097f8758d28e632ff2035e25577f6b060", "tree": "022276b3625a432c7132e39776e7e448445087ac", "parents": [ "488f2eaca1b0831a5a5e6a66e33bad2cdeff7238", "b2d84f078a8be40f5ae3b4d2ac001e2a7f45fe4f" ], "author": { "name": "David Woodhouse", "email": "dwmw2@shinybook.infradead.org", "time": "Thu May 05 13:59:37 2005 +0100" }, "committer": { "name": "David Woodhouse", "email": "dwmw2@shinybook.infradead.org", "time": "Thu May 05 13:59:37 2005 +0100" }, "message": "Merge with master.kernel.org:/pub/scm/linux/kernel/git/torvalds/linux-2.6.git\n" }, { "commit": "a71f62edc935fbdc346c5b16bea19f1480d29635", "tree": "fce9b26a5a9e059735050415eab79fb06a24eae3", "parents": [ "a5a70b75d93b26e14c0c5e759099d602a480b9e2" ], "author": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Tue May 03 16:21:45 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Tue May 03 16:21:45 2005 -0700" }, "message": "[IA64] Fix two warnings introduced by perfmon patches.\n\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "a5a70b75d93b26e14c0c5e759099d602a480b9e2", "tree": "4cd2f4f47cccb8f59e54433ab642e187cbf5e291", "parents": [ "8df5a500a3e97f7811cdce0f553ca1917ccd4220" ], "author": { "name": "stephane eranian", "email": "eranian@hpl.hp.com", "time": "Mon Apr 18 11:42:00 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Tue May 03 15:47:58 2005 -0700" }, "message": "[IA64] another perfmon fix (take2)\n\n- pfm_context_load(): change return value from EINVAL to EBUSY\n when context is already loaded.\n\n- pfm_check_task_state(): pass test if context state is MASKED.\n It is safe to give access on PFM_CTX_MASKED because the PMU\n state (PMD) is stable and saved in software state.\n This helps multiplexing programs such as the example given\n in libpfm-3.1.\n\nSigned-off-by: stephane eranian \u003ceranian@hpl.hp.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "8df5a500a3e97f7811cdce0f553ca1917ccd4220", "tree": "cac08cc58a5e9fb846a9946a52026fcbe7718634", "parents": [ "012914dad25bd5cacf88af4429eecda62a06020d" ], "author": { "name": "Stephane Eranian", "email": "eranian@hpl.hp.com", "time": "Mon Apr 11 13:45:00 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Tue May 03 15:44:48 2005 -0700" }, "message": "[IA64] perfmon \u0026 PAL_HALT again\n\nThe pmu_active test is based on the values of PSR.up. THIS IS THE PROBLEM as\nit does not take into account the lazy restore logic which is as follow (simplified):\n\ncontext switch out:\n\tsave PMDs\n\tclear psr.up\n\trelease ownership\n\ncontext switch in:\n\tif (ctx-\u003elast_cpu \u003d\u003d smp_processor_id() \u0026\u0026 ctx-\u003ecpu_activation \u003d\u003d cpu_activation) {\n\t\tset psr.up\n\t\treturn\n\t}\n\trestore PMD\n\trestore PMC\n\tctx-\u003elast_cpu \u003d smp_processor_id();\n\tctx-\u003eactivation \u003d ++cpu_activation;\n\tset psr.up\n\nThe key here is that on context switch out, we clear psr.up and on context switch in\nwe check if nobody else used the PMU on that processor since last time we came. In\nthat case, we assume the PMD/PMC are ours and we simply reactivate.\n\nThe Caliper problem is that between the moment we context switch out and the moment we\ncome back, nobody effectively used the PMU BUT the processor went idle. Normally this\nwould have no incidence but PAL_HALT does alter the PMU registers. In default_idle(),\nthe test on psr.up is not strong enough to cover this case and we go into PAL which\ntrashed the PMU resgisters. When we come back we falsely assume that this is our state\nyet it is corrupted. Very nasty indeed.\n\nTo avoid the problem it is necessary to forbid going to PAL_HALT as soon as perfmon\ninstalls some valid state in the PMU registers. This happens with an application\nattaches a context to a thread or CPU. It is not enough to check the psr/dcr bits.\nHence I propose the attached patch. It adds a callback in process.c to modify the\ncondition to enter PAL on idle. Basically, now it is conditional to pal_halt\u003d1 AND\nperfmon saying it is okay.\n\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "b1b901c2029aa0fd8aa4ac5f04f31648ae2358b4", "tree": "43f5dde2309d66fa2112a7c2dc46ea7c63fae459", "parents": [ "446b8831f5acf2076fa58a66286789eb84f3df2c" ], "author": { "name": "Russ Anderson", "email": "rja@sgi.com", "time": "Wed Apr 06 00:07:00 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Tue May 03 13:47:42 2005 -0700" }, "message": "[IA64] MCA recovery improvements\n\nJack Steiner uncovered some opportunities for improvement in\nthe MCA recovery code.\n\n 1) Set bsp to save registers on the kernel stack.\n 2) Disable interrupts while in the MCA recovery code.\n 3) Change the way the user process is killed, to avoid \n a panic in schedule.\n\nTesting shows that these changes make the recovery code much \nmore reliable with the 2.6.12 kernel.\n\nSigned-off-by: Russ Anderson \u003crja@sgi.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "446b8831f5acf2076fa58a66286789eb84f3df2c", "tree": "250506863a6e103444584d609cdb577e87b3a499", "parents": [ "7d5f9c0f10255000ca007fb03773c6b825c2b9ce" ], "author": { "name": "David Woodhouse", "email": "dwmw2@infradead.org", "time": "Tue Apr 05 17:47:00 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Tue May 03 13:45:39 2005 -0700" }, "message": "[IA64] fix ia64 syscall auditing\n\nAttached is a patch against David\u0027s audit.17 kernel that adds checks\nfor the TIF_SYSCALL_AUDIT thread flag to the ia64 system call and\nsignal handling code paths. The patch enables auditing of system\ncalls set up via fsys_bubble_down, as well as ensuring that\naudit_syscall_exit() is called on return from sigreturn.\n\nNeglecting to check for TIF_SYSCALL_AUDIT at these points results in\nincorrect information in audit_context, causing frequent system panics\nwhen system call auditing is enabled on an ia64 system.\n\nI have tested this patch and have seen no problems with it.\n\n[Original patch from Amy Griffis ported to current kernel by David Woodhouse]\n\nFrom: Amy Griffis \u003camy.griffis@hp.com\u003e\nFrom: David Woodhouse \u003cdwmw2@infradead.org\u003e\nSigned-off-by: Chris Wright \u003cchrisw@osdl.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "7d5f9c0f10255000ca007fb03773c6b825c2b9ce", "tree": "5f53e41ef4cbe5ced2c5884ee5dde8df1fcaa509", "parents": [ "de7548d0e202263bb6bfd7574a7889e85a691937" ], "author": { "name": "Zwane Mwaikambo", "email": "zwane@arm.linux.org.uk", "time": "Wed Mar 30 21:40:00 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Tue May 03 13:40:18 2005 -0700" }, "message": "[IA64] reduce cacheline bouncing in cpu_idle_wait\n\nAndi noted that during normal runtime cpu_idle_map is bounced around a lot,\nand occassionally at a higher frequency than the timer interrupt wakeup\nwhich we normally exit pm_idle from. So switch to a percpu variable.\n\nI didn\u0027t move things to the slow path because it would involve adding\nscheduler code to wakeup the idle thread on the cpus we\u0027re waiting for.\n\nSigned-off-by: Zwane Mwaikambo \u003czwane@arm.linux.org.uk\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "bb0fc085457cf6a865b8232b0cefab3a7819df44", "tree": "deaa9e3b58d6e40cb2bdba61898bf8373dc02688", "parents": [ "c0b12422e5e1d041026dd27074de17d2d7e32c4e" ], "author": { "name": "Alex Williamson", "email": "alex.williamson@hp.com", "time": "Thu Mar 24 22:58:00 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Tue May 03 13:33:18 2005 -0700" }, "message": "[IA64] use common pxm function\n\nThis patch simplifies a couple places where we search for _PXM\nvalues in ACPI namespace. Thanks,\n\nSigned-off-by: Alex Williamson \u003calex.williamson@hp.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "9df6f705c0934a49d0f0a3468a5b5044c8aec4f1", "tree": "045807d5f55b61d6222c6dada311237a9badeddc", "parents": [ "512f64295f2f0049515dcc4e97c1f1ae0df9629c" ], "author": { "name": "David Mosberger-Tang", "email": "davidm@hpl.hp.com", "time": "Fri Mar 25 00:16:00 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Tue May 03 10:56:42 2005 -0700" }, "message": "[IA64] fix typos caught by new assembler\n\nPatch below fixes 3 trivial typos which are caught by the new\nassembler (v2.169.90). Please apply.\n\n[Note: fix to memcpy that was also part of this patch was separately\n applied from patches by H.J. and Andreas ... so the delta here only\n has the other two fixes. -Tony]\n\nSigned-off-by: David Mosberger-Tang \u003cdavidm@hpl.hp.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "27b030d58c8e72fc7a95187a791bd9406e350f02", "tree": "ab3bab7f39a5ce5bab65578a7e08fa4dfdeb198c", "parents": [ "79d20b14a0d651f15b0ef9a22b6cf12d284a6d38", "6628465e33ca694bd8fd5c3cf4eb7ff9177bc694" ], "author": { "name": "David Woodhouse", "email": "dwmw2@shinybook.infradead.org", "time": "Tue May 03 08:14:09 2005 +0100" }, "committer": { "name": "David Woodhouse", "email": "dwmw2@shinybook.infradead.org", "time": "Tue May 03 08:14:09 2005 +0100" }, "message": "Merge with master.kernel.org:/pub/scm/linux/kernel/git/torvalds/linux-2.6.git\n\n" }, { "commit": "7ed20e1ad521b5f5df61bf6559ae60738e393741", "tree": "90931724e45eaedb3445314e8b94e78253642395", "parents": [ "e5bdd883a189243541e7a132385580703b049102" ], "author": { "name": "Jesper Juhl", "email": "juhl-lkml@dif.dk", "time": "Sun May 01 08:59:14 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Sun May 01 08:59:14 2005 -0700" }, "message": "[PATCH] convert that currently tests _NSIG directly to use valid_signal()\n\nConvert most of the current code that uses _NSIG directly to instead use\nvalid_signal(). This avoids gcc -W warnings and off-by-one errors.\n\nSigned-off-by: Jesper Juhl \u003cjuhl-lkml@dif.dk\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "7d87e14c236d6c4cab66d87cf0bc1e0f0375d308", "tree": "0c0826cdc102286b541e3e56b59c81752d34c90d", "parents": [ "434498d5323445b59167fd7aa5633b74ebbce901" ], "author": { "name": "Stephen Rothwell", "email": "sfr@canb.auug.org.au", "time": "Sun May 01 08:59:12 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Sun May 01 08:59:12 2005 -0700" }, "message": "[PATCH] consolidate sys_shmat\n\nSigned-off-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "3ac3ed555bec5b1f92bb22cb94823a0e99d0f320", "tree": "0133d690f115c47026a316b8b8e2a0a403443d84", "parents": [ "c7fcb0ee74ef4cfdea02befacb55945c93641e44" ], "author": { "name": "Amy Griffis", "email": "amy.griffis@hp.com", "time": "Fri Apr 29 16:12:55 2005 +0100" }, "committer": { "name": "", "email": "dwmw2@shinybook.infradead.org", "time": "Fri Apr 29 16:12:55 2005 +0100" }, "message": "[PATCH] fix ia64 syscall auditing\n\nAttached is a patch against David\u0027s audit.17 kernel that adds checks\nfor the TIF_SYSCALL_AUDIT thread flag to the ia64 system call and\nsignal handling code paths.The patch enables auditing of system\ncalls set up via fsys_bubble_down, as well as ensuring that\naudit_syscall_exit() is called on return from sigreturn.\n\nNeglecting to check for TIF_SYSCALL_AUDIT at these points results in\nincorrect information in audit_context, causing frequent system panics\nwhen system call auditing is enabled on an ia64 system.\n\nSigned-off-by: Amy Griffis \u003camy.griffis@hp.com\u003e\nSigned-off-by: David Woodhouse \u003cdwmw2@infradead.org\u003e\n" }, { "commit": "2fd6f58ba6efc82ea2c9c2630f7ff5ed9eeaf34a", "tree": "87cf236a78ad242ae01f1b71c289131e6d1c0662", "parents": [ "ea3834d9fb348fb1144ad3affea22df933eaf62e" ], "author": { "name": "", "email": "dwmw2@shinybook.infradead.org", "time": "Fri Apr 29 16:08:28 2005 +0100" }, "committer": { "name": "", "email": "dwmw2@shinybook.infradead.org", "time": "Fri Apr 29 16:08:28 2005 +0100" }, "message": "[AUDIT] Don\u0027t allow ptrace to fool auditing, log arch of audited syscalls.\n\nWe were calling ptrace_notify() after auditing the syscall and arguments,\nbut the debugger could have _changed_ them before the syscall was actually\ninvoked. Reorder the calls to fix that.\n\nWhile we\u0027re touching ever call to audit_syscall_entry(), we also make it\ntake an extra argument: the architecture of the syscall which was made,\nbecause some architectures allow more than one type of syscall.\n\nAlso add an explicit success/failure flag to audit_syscall_exit(), for\nthe benefit of architectures which return that in a condition register\nrather than only returning a single register.\n\nChange type of syscall return value to \u0027long\u0027 not \u0027int\u0027.\n\nSigned-off-by: David Woodhouse \u003cdwmw2@infradead.org\u003e\n" }, { "commit": "8e3e50168c8537807b7a6f78588cd72e21363262", "tree": "a63eb89df905efd86a58940286a63c5f670b1c05", "parents": [ "e7e965fa1961a8ce32cbbb1bd436c655ad03973e" ], "author": { "name": "David Mosberger-Tang", "email": "davidm@hpl.hp.com", "time": "Wed Apr 27 21:22:40 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed Apr 27 21:22:40 2005 -0700" }, "message": "[IA64] need r29\u003dpsr *after* rsm psr.i\n\nYanmin Zhang pointed out a sequence problem when saving the psr. David\nMosberger provided this patch (which gave up a cycle).\n\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "e7e965fa1961a8ce32cbbb1bd436c655ad03973e", "tree": "fce517f87be25a0e49d62adb56460a4e7ee36716", "parents": [ "fbf7192ba06e4b55022231e5fdb9db8ce3ad6307" ], "author": { "name": "David Mosberger-Tang", "email": "davidm@hpl.hp.com", "time": "Wed Apr 27 21:22:08 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed Apr 27 21:22:08 2005 -0700" }, "message": "[IA64] use srlz.d instead of srlz.i in ia64_leave_kernel()\n\nThis patch switches the srlz.i in ia64_leave_kernel() to srlz.d. As\nper architecture manual, the former is needed only to ensure that the\nclearing of PSR.IC is seen by the VHPT for subsequent instruction\nfetches. However, since the remainder of the code (up to and\nincluding the RFI instruction) is mapped by a pinned TLB entry, there\nis no chance of an iTLB miss and we don\u0027t care whether or not the VHPT\nsees PSR.IC cleared. Since srlz.d is substantially cheaper than\nsrlz.i, this should shave off a few cycles off the interrupt path\n(unverified though; I\u0027m not setup to measure this at the moment).\n\nSigned-off-by: David Mosberger-Tang \u003cdavidm@hpl.hp.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "fbf7192ba06e4b55022231e5fdb9db8ce3ad6307", "tree": "1da118375605f1a44e7becbd33e0078d27680504", "parents": [ "1ba7be7d691f6df2557d39c5b1a2e14c32e5dd20" ], "author": { "name": "David Mosberger-Tang", "email": "davidm@hpl.hp.com", "time": "Wed Apr 27 21:21:26 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed Apr 27 21:21:26 2005 -0700" }, "message": "[IA64] Annotate fsys_bubble_down() with McKinley dispatch info.\n\nThis patch changes comments \u0026 formatting only. There is no code\nchange.\n\nSigned-off-by: David Mosberger-Tang \u003cdavidm@hpl.hp.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "1ba7be7d691f6df2557d39c5b1a2e14c32e5dd20", "tree": "f6c805c01be475f21de0cdcada8f69c9076ea61e", "parents": [ "21bc4f9b34cc1eab3610955207f72c52495ae8ed" ], "author": { "name": "David Mosberger-Tang", "email": "davidm@hpl.hp.com", "time": "Wed Apr 27 21:20:51 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed Apr 27 21:20:51 2005 -0700" }, "message": "[IA64] Reschedule fsys_bubble_down().\n\nImprovements come from eliminating srlz.i, not scheduling AR/CR-reads\ntoo early (while there are others still pending), scheduling the\nbacking-store switch as well as possible, splitting the BBB bundle\ninto a MIB/MBB pair.\n\nWhy is it safe to eliminate the srlz.i? Observe\nthat we used to clear bits ~PSR_PRESERVED_BITS in PSR.L. Since\nPSR_PRESERVED_BITS\u003d\u003dPSR.{UP,MFL,MFH,PK,DT,PP,SP,RT,IC}, we\nended up clearing PSR.{BE,AC,I,DFL,DFH,DI,DB,SI,TB}. However,\n\n PSR.BE : already is turned off in __kernel_syscall_via_epc()\n PSR.AC : don\u0027t care (kernel normally turns PSR.AC on)\n PSR.I : already turned off by the time fsys_bubble_down gets invoked\n PSR.DFL: always 0 (kernel never turns it on)\n PSR.DFH: don\u0027t care --- kernel never touches f32-f127 on its own\n\t initiative\n PSR.DI : always 0 (kernel never turns it on)\n PSR.SI : always 0 (kernel never turns it on)\n PSR.DB : don\u0027t care --- kernel never enables kernel-level breakpoints\n PSR.TB : must be 0 already; if it wasn\u0027t zero on entry to\n\t __kernel_syscall_via_epc, the branch to fsys_bubble_down\n\t will trigger a taken branch; the taken-trap-handler then\n\t converts the syscall into a break-based system-call.\n\nIn other words: all the bits we\u0027re clearying are either 0 already or\nare don\u0027t cares! Thus, we don\u0027t have to write PSR.L at all and we\ndon\u0027t have to do a srlz.i either.\n\nGood for another ~20 cycle improvement for EPC-based heavy-weight\nsyscalls.\n\nSigned-off-by: David Mosberger-Tang \u003cdavidm@hpl.hp.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "21bc4f9b34cc1eab3610955207f72c52495ae8ed", "tree": "2fc89a932dcc7e0b62717bbf0838136346ebd193", "parents": [ "70929a57cfea8c18de13fcea9ae6771018a98949" ], "author": { "name": "David Mosberger-Tang", "email": "davidm@hpl.hp.com", "time": "Wed Apr 27 21:20:11 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed Apr 27 21:20:11 2005 -0700" }, "message": "[IA64] Annotate __kernel_syscall_via_epc() with McKinley dispatch info.\n\nTwo other very minor changes: use \"mov.i\" instead of \"mov\" for reading\nar.pfs (for clarity; doesn\u0027t affect the code at all). Also, predicate\nthe load of r14 for consistency.\n\nSigned-off-by: David Mosberger-Tang \u003cdavidm@hpl.hp.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "70929a57cfea8c18de13fcea9ae6771018a98949", "tree": "1371e183617f368b7a92b185a2dee829c70d0efd", "parents": [ "f8fa5448fc9b4a7806b1297a0b57808f12fe4d43" ], "author": { "name": "David Mosberger-Tang", "email": "davidm@hpl.hp.com", "time": "Wed Apr 27 21:19:37 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed Apr 27 21:19:37 2005 -0700" }, "message": "[IA64] Reschedule __kernel_syscall_via_epc().\n\nAvoid some stalls, which is good for about 2 cycles when invoking a\nlight-weight handler. When invoking a heavy-weight handler, this\nhelps by about 7 cycles, with most of the improvement coming from the\nimproved branch-prediction achieved by splitting the BBB bundle into\ntwo MIB bundles.\n\nSigned-off-by: David Mosberger-Tang \u003cdavidm@hpl.hp.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "f8fa5448fc9b4a7806b1297a0b57808f12fe4d43", "tree": "b32f9939ee3b72010e4c3255d58278f85c547a7e", "parents": [ "c03f058fbf685f2ff630095d2c1e98d331b81e82" ], "author": { "name": "David Mosberger-Tang", "email": "davidm@hpl.hp.com", "time": "Wed Apr 27 21:19:04 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed Apr 27 21:19:04 2005 -0700" }, "message": "[IA64] Reschedule break_fault() for better performance.\n\nThis patch reorganizes break_fault() to optimistically assume that a\nsystem-call is being performed from user-space (which is almost always\nthe case). If it turns out that (a) we\u0027re not being called due to a\nsystem call or (b) we\u0027re being called from within the kernel, we fixup\nthe no-longer-valid assumptions in non_syscall() and .break_fixup(),\nrespectively.\n\nWith this approach, there are 3 major phases:\n\n - Phase 1: Read various control \u0026 application registers, in\n\t particular the current task pointer from AR.K6.\n - Phase 2: Do all memory loads (load system-call entry,\n\t load current_thread_info()-\u003eflags, prefetch\n\t kernel register-backing store) and switch\n\t to kernel register-stack.\n - Phase 3: Call ia64_syscall_setup() and invoke\n\t syscall-handler.\n\nGood for 26-30 cycles of improvement on break-based syscall-path.\n\nSigned-off-by: David Mosberger-Tang \u003cdavidm@hpl.hp.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "c03f058fbf685f2ff630095d2c1e98d331b81e82", "tree": "def1271117756fdde22b4d51b03a43378ad927ca", "parents": [ "87e522a0f7f8a7a5a1a880517989835c4f09c63e" ], "author": { "name": "David Mosberger-Tang", "email": "davidm@hpl.hp.com", "time": "Wed Apr 27 21:18:22 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed Apr 27 21:18:22 2005 -0700" }, "message": "[IA64] In ia64_leave_syscall(), fix comments and whitespace only.\n\nSigned-off-by: David Mosberger-Tang \u003cdavidm@hpl.hp.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "87e522a0f7f8a7a5a1a880517989835c4f09c63e", "tree": "72a20dd374169677a3b151f1f1168e2654cecc08", "parents": [ "060561ff79b01eea58e6d72abfb8e7580ff21f2a" ], "author": { "name": "David Mosberger-Tang", "email": "davidm@hpl.hp.com", "time": "Wed Apr 27 21:17:44 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed Apr 27 21:17:44 2005 -0700" }, "message": "[IA64] Schedule ia64_leave_syscall() to read ar.bsp earlier\n\nReschedule code to read ar.bsp as early as possible. To enable this,\ndon\u0027t bother clearing some of the registers when we\u0027re returning to\nkernel stacks. Also, instead of trying to support the pNonSys case\n(which makes no sense), do a bugcheck instead (with break 0). Finally,\nremove a clear of r14 which is a left-over from the previous patch.\n\nSigned-off-by: David Mosberger-Tang \u003cdavidm@hpl.hp.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "060561ff79b01eea58e6d72abfb8e7580ff21f2a", "tree": "7c3f95456480d3c6b44842df5a7619be98f2e14d", "parents": [ "96e017495e6833adcbad84783e3c5eb685836bbf" ], "author": { "name": "David Mosberger-Tang", "email": "davidm@hpl.hp.com", "time": "Wed Apr 27 21:17:03 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed Apr 27 21:17:03 2005 -0700" }, "message": "[IA64] In syscall-entry, use st8 instead of stf8 to clear pt_regs.r8\n\nUsing stf8 seemed like a clever idea at the time, but stf8 forces\nthe cache-line to be invalidated in the L1D (if it happens to be\nthere already). This patch eliminates a guaranteed L1D cache-miss\nand, by itself, is good for a 1-2 cycle improvement for heavy-weight\nsyscalls.\n\nSigned-off-by: David Mosberger-Tang \u003cdavidm@hpl.hp.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "96e017495e6833adcbad84783e3c5eb685836bbf", "tree": "41504d1ba4f94282d109224936ae6829109c3e6d", "parents": [ "3c79c8b1d92a9ae3edf3cbcd2c43c553ee0f1d83" ], "author": { "name": "David Mosberger-Tang", "email": "davidm@hpl.hp.com", "time": "Wed Apr 27 21:16:07 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed Apr 27 21:16:07 2005 -0700" }, "message": "[IA64] On return from syscall, hint b7 with __kernel_syscall_via_epc().\n\nWhy is this a good idea? Clearing b7 to 0 is guaranteed to do us no\ngood and writing it with __kernel_syscall_via_epc() yields a 6 cycle\nimprovement _if_ the application performs another EPC-based system-\ncall without overwriting b7, which is not all that uncommon. Well\nworth the minimal cost of 1 bundle of code.\n\nSigned-off-by: David Mosberger-Tang \u003cdavidm@hpl.hp.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "3c79c8b1d92a9ae3edf3cbcd2c43c553ee0f1d83", "tree": "50a6d59383a15d92aa054371a247112e3e539bc5", "parents": [ "9ec1a7ad434fa951ff845dbee3397cf6ad4f76df" ], "author": { "name": "David Mosberger-Tang", "email": "davidm@hpl.hp.com", "time": "Wed Apr 27 21:15:13 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed Apr 27 21:15:13 2005 -0700" }, "message": "[IA64] Schedule fp-clearing insns at least 6 cycles after reading ar.bsp.\n\nDecreases syscall overhead by approximately 6 cycles.\n\nSigned-off-by: David Mosberger-Tang \u003cdavidm@hpl.hp.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "9ec1a7ad434fa951ff845dbee3397cf6ad4f76df", "tree": "05e8a156ccd42bfbed3293b386f0e4d905d04a88", "parents": [ "06ef66081620e94fe35a518f98624b83a140096e" ], "author": { "name": "David Mosberger-Tang", "email": "davidm@hpl.hp.com", "time": "Wed Apr 27 21:13:33 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed Apr 27 21:13:33 2005 -0700" }, "message": "[IA64] Use dynamic prediction for RSE-clearing branches.\n\nThis by itself is good for a 1-2 cycle speed up. Effect is bigger\nwhen combined with the later patches.\n\nSigned-off-by: David Mosberger-Tang \u003cdavidm@hpl.hp.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "06ef66081620e94fe35a518f98624b83a140096e", "tree": "a34e95f00e5994b5c5f179637d9047c3ee212269", "parents": [ "fc67b16ecaf6ebde04096030c268adddade023f1" ], "author": { "name": "David Mosberger-Tang", "email": "davidm@hpl.hp.com", "time": "Wed Apr 27 21:10:45 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Wed Apr 27 21:10:45 2005 -0700" }, "message": "[IA64] __ia64_syscall() is no longer used anywhere in the kernel. Remove it.\n\nSigned-off-by: David Mosberger-Tang \u003cdavidm@hpl.hp.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "b9e41d7fb62ae26adee84c18048037214ce5d866", "tree": "d39f224bce0113f860417f81acceb5d2cb8428b3", "parents": [ "e1ed81ab7a34fc0f92f2e200825bdb6d86d6c8ef" ], "author": { "name": "Kenji Kaneshige", "email": "kaneshige.kenji@jp.fujitsu.com", "time": "Mon Apr 25 13:27:48 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Mon Apr 25 13:27:48 2005 -0700" }, "message": "[IA64] iosapic.c: typo ... s/spin_unlock_irq/spin_unlock/\n\nvector sharing patch had a typo ... mismatched spin_lock() with\na spin_unlock_irq(). Fix from Kenji Kaneshige.\n\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "e1ed81ab7a34fc0f92f2e200825bdb6d86d6c8ef", "tree": "488b22465b9108142b2df16d6e267b0ddda2b839", "parents": [ "24eeb568aeeaee771b9f0a6fd6f5d01040a887da" ], "author": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Mon Apr 25 13:27:12 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Mon Apr 25 13:27:12 2005 -0700" }, "message": "[IA64] print \"siblings\" before {physical,core,thread} id\n\nRohit and Suresh changed their mind about the order to print things\nin /proc/cpuinfo, but didn\u0027t include the change in the version of\nthe patch they sent to me.\n\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "24eeb568aeeaee771b9f0a6fd6f5d01040a887da", "tree": "4d3e731845cde292b5a40ff7db97cce639dc073c", "parents": [ "e927ecb05e1ce4bbb1e10f57008c94994e2160f5" ], "author": { "name": "Kenji Kaneshige", "email": "kaneshige.kenji@jp.fujitsu.com", "time": "Mon Apr 25 13:26:23 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Mon Apr 25 13:26:23 2005 -0700" }, "message": "[IA64] vector sharing (Large I/O system support)\n\nCurrent ia64 linux cannot handle greater than 184 interrupt sources\nbecause of the lack of vectors. The following patch enables ia64 linux\nto handle greater than 184 interrupt sources by allowing the same\nvector number to be shared by multiple IOSAPIC\u0027s RTEs. The design of\nthis patch is besed on \"Intel(R) Itanium(R) Processor Family Interrupt\nArchitecture Guide\".\n\nEven if you don\u0027t have a large I/O system, you can see the behavior of\nvector sharing by changing IOSAPIC_LAST_DEVICE_VECTOR to fewer value.\n\nSigned-off-by: Kenji Kaneshige \u003ckaneshige.kenji@jp.fujitsu.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "e927ecb05e1ce4bbb1e10f57008c94994e2160f5", "tree": "bc8256cc074f02d557088696035982fa7ae0b301", "parents": [ "6118ec847e8e35393efc0f88394c2f5dd48c3313" ], "author": { "name": "Suresh Siddha", "email": "suresh.b.siddha@intel.com", "time": "Mon Apr 25 13:25:06 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Mon Apr 25 13:25:06 2005 -0700" }, "message": "[IA64] multi-core/multi-thread identification\n\nVersion 3 - rediffed to apply on top of Ashok\u0027s hotplug cpu\npatch. /proc/cpuinfo output in step with x86.\n\nThis is an updated MC/MT identification patch based on the \nprevious discussions on list. \n\nAdd the Multi-core and Multi-threading detection for IPF.\n - Add new core and threading related fields in /proc/cpuinfo.\n\t\tPhysical id\n\t\tCore id\n\t\tThread id\n\t\tSiblings\n - setup the cpu_core_map and cpu_sibling_map appropriately\n - Handles Hot plug CPU\n \nSigned-off-by: Suresh Siddha \u003csuresh.b.siddha@intel.com\u003e\nSigned-off-by: Gordon Jin \u003cgordon.jin@intel.com\u003e\nSigned-off-by: Rohit Seth \u003crohit.seth@intel.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "a37d98f6a98254c05315e0bbf45c4602942d14b1", "tree": "8d6a3b98118866319a76d719efa9d7fbe5914854", "parents": [ "93a07d0a0e7b013ee73fb39d4edb07b47288912e" ], "author": { "name": "David Mosberger-Tang", "email": "davidm@hpl.hp.com", "time": "Mon Apr 25 13:20:38 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Mon Apr 25 13:20:38 2005 -0700" }, "message": "[IA64] fix syscall-optimization goof\n\nSadly, I goofed in this syscall-tuning patch:\n\nChangeSet 1.1966.1.40 2005/01/22 13:31:05 davidm@hpl.hp.com\n [IA64] Improve ia64_leave_syscall() for McKinley-type cores.\n\n Optimize ia64_leave_syscall() a bit better for McKinley-type cores.\n The patch looks big, but that\u0027s mostly due to renaming r16/r17 to r2/r3.\n Good for a 13 cycle improvement.\n\nThe problem is that the size of the physical stacked registers was\nloaded into the wrong register (r3 instead of r17). Since r17 by\ncoincidence always had the value 1, this had the effect of turning\nrse_clear_invalid into a no-op. That poses the risk of leaking kernel\nstate back to user-land and is hence not acceptable.\n\nThe fix below is simple, but unfortunately it costs us about 28 cycles\nin syscall overhead. ;-(\n\nUnfortunately, there isn\u0027t much we can do about that since those\nregisters have to be cleared one way or another.\n\n\t--david\n\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "4944930ab748942e41ea4dc313fcb0946aee3f17", "tree": "5d23e420c0d7f4387237028ca2dca1ec97f59861", "parents": [ "658b32cad9ae087bd34f35a925fd75b76d663d4e" ], "author": { "name": "Stephane Eranian", "email": "eranian@hpl.hp.com", "time": "Mon Apr 25 13:08:30 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Mon Apr 25 13:08:30 2005 -0700" }, "message": "[IA64] perfmon: make pfm_sysctl a global, and other cleanup\n\n- make pfm_sysctl a global such that it is possible\n to enable/disable debug printk in sampling formats\n using PFM_DEBUG.\n\n- remove unused pfm_debug_var variable\n\n- fix a bug in pfm_handle_work where an BUG_ON() could\n be triggered. There is a path where pfm_handle_work()\n can be called with interrupts enabled, i.e., when\n TIF_NEED_RESCHED is set. The fix correct the masking\n and unmasking of interrupts in pfm_handle_work() such\n that we restore the interrupt mask as it was upon entry.\n\nsigned-off-by: stephane eranian \u003ceranian@hpl.hp.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "30325d17715302a60c9afdaacaafaeb056b7e880", "tree": "8872d59f841a5338e04f15c1183b8fd2ad4bfbb3", "parents": [ "e8d1cb2f280aa53e1c75c8b5fcbf80b3481d0caa" ], "author": { "name": "David Mosberger-Tang", "email": "davidm@hpl.hp.com", "time": "Mon Apr 25 13:03:16 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Mon Apr 25 13:03:16 2005 -0700" }, "message": "[IA64] speed up syscall path a bit more\n\nRecently I noticed that clearing ar.ssd/ar.csd right before srlz.d is\ncausing significant stalling in the syscall path. The patch below\nfixes that by moving the register-writes after srlz.d. On a Madison,\nthis drops break-based getpid() from 241 to 226 cycles (-15 cycles).\n\nSigned-off-by: David Mosberger-Tang \u003cdavidm@hpl.hp.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "e8d1cb2f280aa53e1c75c8b5fcbf80b3481d0caa", "tree": "832d5f4beabfc1c075dd54eebe343ad10c611c73", "parents": [ "8297511530b0d2a281c796e738683951a59a020c" ], "author": { "name": "Keith Owens", "email": "kaos@sgi.com", "time": "Mon Apr 25 11:45:26 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Mon Apr 25 11:45:26 2005 -0700" }, "message": "[IA64] Tighten up unw_unwind_to_user check\n\nDetect user space by the unwind frame with predicate PRED_USER_STACK\nset, instead of a user space IP. Tighten up the last ditch check for\nrunning off the top of the kernel stack.\n\nBased on a suggestion by David Mosberger, reworked to fit the current\ntree. This survives my stress test which used to break 2.6.9 kernels.\nUnlike 2.6.11, the stress test now unwinds to the correct point, so\ngdb can get the user space registers.\n\nSigned-off-by: Keith Owens \u003ckaos@sgi.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "8297511530b0d2a281c796e738683951a59a020c", "tree": "3ade89d674db6b0791ca72704478214d0bb56ad1", "parents": [ "bf1cf98fa941fea5e630e341db4a294d531aaa3e" ], "author": { "name": "David Mosberger-Tang", "email": "davidm@hpl.hp.com", "time": "Mon Apr 25 11:44:02 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Mon Apr 25 11:44:02 2005 -0700" }, "message": "[IA64] add missing cpu_relax() in ITC syncing code\n\nCall cpu_relax() in busy-waiting loops of the ITC-syncing code.\n\nSigned-off-by: David Mosberger-Tang \u003cdavidm@hpl.hp.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "df6c6804ce1d383c878ab071bce8125125d96abc", "tree": "43765d4b40a4a3c64f2a32b684f4041e2f01644d", "parents": [ "b8d8b883e6f029e99c35c88f853501740e322131" ], "author": { "name": "Ashok Raj", "email": "ashok.raj@intel.com", "time": "Fri Apr 22 14:46:24 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Fri Apr 22 14:46:24 2005 -0700" }, "message": "[IA64] Fix build errors for !HOTPLUG case.\n\nSigned-off-by: Ashok Raj \u003cashok.raj@intel.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "b8d8b883e6f029e99c35c88f853501740e322131", "tree": "391f2ade8823149f217991eb02911bf3dacce050", "parents": [ "7130667107cd3ab9d6802b69bab63c7d22f20bd4" ], "author": { "name": "Ashok Raj", "email": "ashok.raj@intel.com", "time": "Fri Apr 22 14:44:40 2005 -0700" }, "committer": { "name": "Tony Luck", "email": "tony.luck@intel.com", "time": "Fri Apr 22 14:44:40 2005 -0700" }, "message": "[IA64] cpu hotplug: return offlined cpus to SAL\n\nThis patch is required to support cpu removal for IPF systems. Existing code\njust fakes the real offline by keeping it run the idle thread, and polling\nfor the bit to re-appear in the cpu_state to get out of the idle loop.\n\nFor the cpu-offline to work correctly, we need to pass control of this CPU \nback to SAL so it can continue in the boot-rendez mode. This gives the\nSAL control to not pick this cpu as the monarch processor for global MCA\nevents, and addition does not wait for this cpu to checkin with SAL\nfor global MCA events as well. The handoff is implemented as documented in \nSAL specification section 3.2.5.1 \"OS_BOOT_RENDEZ to SAL return State\"\n\nSigned-off-by: Ashok Raj \u003cashok.raj@intel.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n" }, { "commit": "1da177e4c3f41524e886b7f1b8a0c1fc7321cac2", "tree": "0bba044c4ce775e45a88a51686b5d9f90697ea9d", "parents": [], "author": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Sat Apr 16 15:20:36 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Sat Apr 16 15:20:36 2005 -0700" }, "message": "Linux-2.6.12-rc2\n\nInitial git repository build. I\u0027m not bothering with the full history,\neven though we have it. We can create a separate \"historical\" git\narchive of that later if we want to, and in the meantime it\u0027s about\n3.2GB when imported into git - space that would just make the early\ngit days unnecessarily complicated, when we don\u0027t have a lot of good\ninfrastructure for it.\n\nLet it rip!\n" } ] }